Boyle PA, et al. Person-specific contribution of neuropathologies to cognitive loss in old age. Ann Neurol. 2018;83:74-83. PMID 29244218¶
One-paragraph summary¶
Participants (n=1,079) from two longitudinal clinical-pathological cohort studies of ageing completed two or more cognitive evaluations (maximum 22), died, and underwent neuropathologic examination for Alzheimer's disease, other neurodegenerative diseases and vascular pathologies. Linear mixed models estimated the proportion of each individual's cognitive loss attributable to each of nine neuropathologies. Neuropathology was near-ubiquitous: 94% had at least one, 78% at least two, 58% at least three and 35% at least four. Alzheimer's disease was the most frequent single pathology (65%) but occurred in isolation in only 9%. More than 230 different neuropathologic combinations were observed, each in fewer than 6% of the cohort. Alzheimer's disease accounted for about 50% of the observed cognitive loss on average, but the person-specific proportion ranged from 22% to 100%. Lewy bodies and hippocampal sclerosis also had potent, person-variable effects (PMID 29244218).
Key findings¶
| Quantity | Value |
|---|---|
| ≥1 neuropathology | 94% |
| ≥2 | 78% |
| ≥3 | 58% |
| ≥4 | 35% |
| AD present | 65% |
| AD in isolation | 9% |
| Distinct pathology combinations | >230, each in <6% |
| AD share of cognitive loss, mean | ~50% |
| AD share of cognitive loss, person-specific range | 22%–100% |
Limitations¶
- Two research cohorts with high consent-to-autopsy rates; participants are volunteers with above-average education and follow-up, so the mixture may not generalise to all populations.
- Attribution is model-based: the proportion of cognitive loss assigned to each pathology is an estimate from a linear mixed model, not a measured causal quantity.
- Nine pathologies were assessed; anything unmeasured (including pathologies not yet defined at the time) is absorbed into the residual.
- Autopsy is a single terminal snapshot of a lifetime process.
Why it matters¶
This paper converts "mixed pathology is common" from a qualitative observation (Schneider 2007, PMID 17568013) into a distribution: there is no modal patient. If Alzheimer's disease explains between 22% and 100% of an individual's cognitive decline and the specific pathology combination is nearly unique to each person, then a single-target therapy addresses a variable and unmeasured fraction of the process producing symptoms — which is the structural reason effect sizes in unselected populations should be expected to be smaller than in biomarker-pure trial populations, and the strongest available argument for combination approaches and for in-vivo markers of non-AD copathology. The authors' own conclusion is that the finding creates "an urgent need for novel therapeutic approaches that embrace the complexity of disease."
Cited by wiki pages¶
- vascular and metabolic contributions
- overview